Skip to content

Onsa MCP server

ai.onsa/onsaonsa.aistreamable-httpv1.0.0Scanned

Find scored B2B leads, read campaign replies and send approved LinkedIn outreach.

16 tools: 6 can delete or cancel things and 10 only read.

Tools: 6 delete, 0 irreversible, 0 write, 10 read.
  • Reads private data · 3
  • Reads text others write · 8
  • Can send data out · 2
  • 4 findings

Tools listed without sign-in. Remote posture

With Anlyon

6 tools here can delete or cancel things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Onsa MCP server can do

The Onsa MCP server is published by onsa.ai, a domain the MCP Registry verified. It has 16 tools. It can delete or cancel things (6 tools: create_shortlist, revoke_shortlist, find_leads, continue_campaign and 2 more). 10 tools only read.

3 tools read private data (list_pending_outreach, list_replies and get_campaign_stats); 8 tools bring in text other people can write (create_shortlist, revoke_shortlist, find_leads and 5 more); 2 tools can send data out of your system (continue_campaign and send_outreach). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 4 findings (credentials-mention, annotation-conflict and long-description). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 11, 2026, version 1.0.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

16 tools

Deletes or cancels 6Claude Code: deny
  • continue_campaignSends an instruction to the agent inside an existing campaign and returns a jobId to read with fetch_leads.Reads text others writeCan send data out1 finding

    Why it is in this group

    • The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe description's first verb is "send": it sends or publishes something to other people.

    Findings

    • Info

      The input schema at properties.taskContext.description mentions "credentials".

      Matched text

      inputSchema.properties.taskContext.description: "...er profile, unrelated personal details, credentials, or guesses. This background does not a..."

      What this check looks for
    Description, as the server wrote it
    Sends an instruction to the agent inside an existing campaign and returns a jobId to read with fetch_leads. This is the tool that grows or steers a cohort in place - 'find 5 more like these', 'look at Singapore and the Gulf instead of US institutions', 'focus on funds over $5bn AuM'. find_leads always creates a separate campaign with its own ICP, which splits the funnel and leaves the two cohorts incomparable. The agent sees the campaign's existing leads and ICP, so they can be referred to. It cannot answer back through this API, so a question sent here gets no response. New leads count against the prospect allowance, and de-duplication is per workspace, so a request for 5 more can yield fewer when the agent rediscovers people already in the workspace. It does not remove or skip leads: 'drop the bad ones' takes nothing out of the cohort or out of the outreach queue.

    Annotations the server declares

    • title: Steer a campaign
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • create_shortlistPublishes a preview_shortlist snapshot as a link.Reads text others writeNo findings

    Why it is in this group

    • The name "create_shortlist" says "create": a word for a change.
    • The description's first verb is "publish": a word for a change that another call does not undo.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Publishes a preview_shortlist snapshot as a link. It requires the previewKey that preview_shortlist returned, and confirmPublic: true. The server itself requests no confirmation from the user: confirmPublic: true records the caller's statement that the user reviewed this snapshot and confirmed publishing it. The server rebuilds the snapshot and creates nothing when the selection or the snapshot no longer matches that previewKey: a different campaign, selection or order of leadIds, a different note, an includeRationale value that changes the snapshot, a prospect changed or removed, or the campaign renamed since the preview. The link makes the snapshot viewable by anyone who has it for 30 days unless it is revoked with revoke_shortlist. This does not share the conversation or invite anyone into the workspace. Returns the URL, list id and expiry to the requester; sending it to another person is a separate user action.

    Annotations the server declares

    • title: Create a public list link
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • find_leadsStarts a live B2B lead search with Onsa's agent, matching real people (with LinkedIn profiles) against the workspace's ICP.Reads text others write2 findings

    Why it is in this group

    • The name "find_leads" says "find": a word for reading.
    • The description's first verb is "start": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The input schema at properties.taskContext.description mentions "credentials".

      Matched text

      inputSchema.properties.taskContext.description: "...er profile, unrelated personal details, credentials, or guesses. This background does not a..."

      What this check looks for
    • Warn

      The server marks "find_leads" destructive (destructiveHint: true), and its name says "find", a word for reading.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "find_leads" says "find": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Starts a live B2B lead search with Onsa's agent, matching real people (with LinkedIn profiles) against the workspace's ICP. Takes a natural-language brief - titles, company type, geography, e.g. 'find 5 fintech founders in NYC'. Returns a jobId immediately, with campaignUrl and a `progress` note. The search itself runs in the background and keeps running after the conversation moves on. First leads arrive within 5 minutes in half of searches and within 9 minutes in 9 of 10. More can arrive until the final report, typically about 10 minutes after the start (9 of 10 within 15). Its status and results are read with fetch_leads, with the same jobId at any later time. Hosts that render MCP Apps show a live card that follows the search and lists leads as they arrive. `limit` is a target the agent aims at rather than a cap, so it often returns more than asked. Every lead it finds counts against the workspace's prospect allowance.

    Annotations the server declares

    • title: Find leads
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • revoke_shortlistRevokes a list link by the id create_shortlist returned; only the user who created the link can revoke it.Reads text others writeNo findings

    Why it is in this group

    • The name "revoke_shortlist" says "revoke": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Revokes a list link by the id create_shortlist returned; only the user who created the link can revoke it. Future views stop immediately; previously copied or downloaded information cannot be recalled.

    Annotations the server declares

    • title: Revoke a list link
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: true
  • rewrite_outreachReplaces the text of an outreach draft that is waiting for approval.No findings

    Why it is in this group

    • The description's first verb is "replace": a word for a change.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Replaces the text of an outreach draft that is waiting for approval. The current draft and the lead's scoreExplanation come from list_pending_outreach; get_lead_memo carries the richer research on that person. The rewritten draft stays in the approval queue, and this tool sends nothing. LinkedIn follow-up drafts (kind: follow_up) cannot be rewritten here: they are edited in Onsa, or approved as written with send_outreach.

    Annotations the server declares

    • title: Rewrite outreach draft
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • send_outreachQueues one already-approved outreach draft for delivery to a real person on LinkedIn.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "send_outreach" says "send": a word for sending, and what is sent cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "send_outreach" says "send": it sends or publishes something to other people.
    Description, as the server wrote it
    Queues one already-approved outreach draft for delivery to a real person on LinkedIn. It requires `confirmText`, the draft body character-for-character as stored, and `confirmName`, the recipient's name: drafts are often near-identical between people, so matching the body alone does not identify which one was meant. A mismatch is refused without returning the stored text, which list_pending_outreach supplies. The server additionally requires a confirmation from the person at the keyboard, rendered by the MCP client and quoting the draft as stored; that approval is single-use and bound to one recipient and one draft. A client that cannot render such a confirmation receives a refusal carrying a link to approve inside the Onsa app, and nothing is queued. On success the message is queued rather than delivered: Onsa sends it on its own schedule, subject to daily pacing limits. A follow-up draft (kind: follow_up) goes into the same LinkedIn thread as the first message and is cancelled if the person replies before it is sent.

    Annotations the server declares

    • title: Send outreach message
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
Only reads 10Show 10 read-only tools
  • fetch_leadsReturns the status and any results of a find_leads or continue_campaign job, by jobId.1 finding

    Why it is in this group

    • The name "fetch_leads" says "fetch": a word for reading.

    Findings

    • Info

      The description is 3,755 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...size; each entry is a suggestion the user can accept, not something already done."

      What this check looks for
    Description, as the server wrote it
    Returns the status and any results of a find_leads or continue_campaign job, by jobId. Status values are "pending", "completed" and "stalled", and `leads` can hold a partial list while "pending". For a find_leads job: "pending" while it has no leads, the agent has not answered and it is under about 15 minutes old, and also while it has leads, until a reply carrying leads completes it; "completed" once the agent posts a reply carrying leads while this is the oldest open job on its campaign, usually its final report (`total` can still change afterwards); "stalled" as soon as the agent answers without any leads (a question, or none found), or when no lead has arrived after about 15 minutes - derived on each read, so a lead that lands later turns it back to "pending". For a continue_campaign job: "pending" while `newLeads` is 0 and it is under about 15 minutes old, even after the agent has replied (if that reply confirms a settings-only change, the request is done), and while `newLeads` is above 0, until a reply carrying leads completes it; "completed" once a reply carrying leads arrives on the campaign while this is the oldest open job there (possibly a late reply from an earlier search), or after about 15 minutes with `newLeads` 0 when an agent reply was confirmed; "stalled" after about 15 minutes with `newLeads` 0 when no agent reply was confirmed (agentMessage can still hold one). A reply carrying leads completes only the oldest open job on a campaign, so a newer job stays "pending" meanwhile, even with `newLeads` above 0. A pending search keeps running when the conversation ends, and a later call with the same jobId returns what it found. First leads arrive within 5 minutes in half of searches and within 9 minutes in 9 of 10. More can arrive until the final report, typically about 10 minutes after the start (9 of 10 within 15). Also returns `total` (for a find_leads job, the leads the campaign holds now, skipped and deleted ones excluded; for a continuation, equal to `newLeads`), `newLeads` (continuations only: leads added to the campaign since the continuation started - a count by time, which can include a late batch from an earlier search), `returned` (how many this response carries), `campaignId` (accepted by get_campaign, get_campaign_leads and get_campaign_stats) and `campaignUrl`, a deep link to the prospects tab for this search in Onsa. `agentMessage` is the latest agent text in this campaign's chat. For a continuation it is filtered to what was said after the continuation started, though it can be a late message from an earlier search; for a find_leads job it is not filtered, so after a later continue_campaign on the same campaign it can be that request’s reply. A null `agentMessage` does not mean the agent is silent: artifact-only messages carry no text, and an agent that errored writes nothing there at all. The agent cannot be replied to through this API. While a search is running, one call can wait up to 35 seconds for its first leads or its next batch before answering, so a call can take that long; other calls answer at once. `progress` carries startedAt, elapsedSeconds and a note on where the search stands, whether the next call will wait, and when results are likely. Each lead carries name, companyName, linkedInUrl, position, headline, location, industry, companyUrl, email (often null), and score (1-5) with scoreExplanation, the reasoning for why this person matches the ICP. A lead has `signal` (status verified, inferred or none, plus a one-sentence summary) only when the search asked for a need signal. `refinements` appears when leads are returned and the campaign does not filter by company size; each entry is a suggestion the user can accept, not something already done.

    Annotations the server declares

    • title: Fetch search results
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_campaignReturns one campaign's ICP - the ideal-customer profile the agent derived and scores leads against - plus its outreach template and settings.No findings

    Why it is in this group

    • The name "get_campaign" says "get": a word for reading.
    Description, as the server wrote it
    Returns one campaign's ICP - the ideal-customer profile the agent derived and scores leads against - plus its outreach template and settings. The ICP comes back exactly as stored, in snake_case: `perfect_lead` and `reachable_market` are one-line summaries, while `company` and `person` hold the rules that actually drive scoring, each an object with `critical` and `preferential` rule lists. `product` and `owner` describe the seller. The two summary strings are not the scoring criteria; `company.critical` and `person.critical` are. No key is guaranteed present. `outreachTemplate` shows how much personalization the messages allow: a template whose only placeholders are [FIRST_NAME] and [COMPANY_NAME] produces near-identical mail-merge copy for every lead.

    Annotations the server declares

    • title: Get campaign
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_campaign_leadsReturns the leads of any campaign by campaignId, with the same fields as fetch_leads, including score and scoreExplanation.No findings

    Why it is in this group

    • The name "get_campaign_leads" says "get": a word for reading.
    Description, as the server wrote it
    Returns the leads of any campaign by campaignId, with the same fields as fetch_leads, including score and scoreExplanation. It covers campaigns not started in this session, which fetch_leads cannot reach because fetch_leads requires a jobId from a find_leads call in the same session. Passing `leadIds` resolves the reply buckets from get_campaign_stats back into named people. A lead carries `signal` only when the search asked for a need signal.

    Annotations the server declares

    • title: Get campaign leads
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_campaign_statsReturns the outreach funnel for one campaign: invites sent, invites accepted, messages sent, and replies split into positive / negative / other by sentiment.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_campaign_stats" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "message": it reads messages.
    • untrusted-inputThe description's first sentence says "return" and "message": it reads messages, which other people can write.
    Description, as the server wrote it
    Returns the outreach funnel for one campaign: invites sent, invites accepted, messages sent, and replies split into positive / negative / other by sentiment. LinkedIn and email are merged, as on Onsa's Overview page. These count leads rather than actions - a lead invited twice counts once - so where a lead was re-invited they read slightly lower than the Overview widget, which counts actions. `invitesSent` can exceed `leadsTotal` without anyone having been invited twice: the counters are already de-duplicated by lead, and the gap means a lead who was contacted has since been skipped or deleted, which action rows survive and leadsTotal does not count. Onsa stores no post likes or emoji reactions at all, so 'LinkedIn reactions' in this data means the replies people sent; list_replies returns their text, and this tool only counts them. `rates` carries acceptancePct, replyPct, positivePct and negativePct, each already computed over its own correct denominator; leadsTotal is not one of those denominators, since it counts every prospect in the cohort including those never contacted. A rate is null when its denominator is 0, meaning nothing was sent so no rate exists - which is different from the counts above being genuinely 0. All four rates count only replies Onsa has scored, so an unscored or still-in-window reply appears in none of them, and list_replies can legitimately show more replies than the rates imply. Two further properties of the data: the *Scheduled counts are everything queued regardless of date, and sentiment is evaluated once per lead ever rather than once per reply. A campaign whose invites were never sent reads as all zeros, which is 'not tried yet' rather than 'failed'.

    Annotations the server declares

    • title: Get campaign funnel
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_lead_memoReturns the research memo Onsa's agent wrote about one lead: role history, company size and stage, what they have said publicly, and the angle on them.No findings

    Why it is in this group

    • The name "get_lead_memo" says "get": a word for reading.
    Description, as the server wrote it
    Returns the research memo Onsa's agent wrote about one lead: role history, company size and stage, what they have said publicly, and the angle on them. It is usually far richer than scoreExplanation, and it is the source material for outreach built on a specific, checkable fact rather than a generic opener. Most leads have no memo - Onsa writes one only for prospects it has researched - so `memo: null` is the common case and not an error, and scoreExplanation is the remaining source in that case.

    Annotations the server declares

    • title: Get lead research memo
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_campaignsLists the campaigns (past lead searches) in this workspace that the user takes part in, newest first.No findings

    Why it is in this group

    • The name "list_campaigns" says "list": a word for reading.
    Description, as the server wrote it
    Lists the campaigns (past lead searches) in this workspace that the user takes part in, newest first. Returns the newest `limit` of them, default 50; `returned` against `total` shows whether older campaigns were omitted. Each entry has id, title, leadsTotal, hasIcp, tags, createdAt and updatedAt. An id is accepted by get_campaign for its ICP, get_campaign_leads for its people, get_campaign_stats for its outreach funnel, list_replies for what prospects wrote back, and list_next_steps for what the campaign still needs a human to do. Searches started over MCP often share a generic title, so the ICP and the dates distinguish cohorts more reliably than the title alone. A campaign row exists from the moment a search starts, so the newest entry is frequently still empty, with leadsTotal 0.

    Annotations the server declares

    • title: List campaigns
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_next_stepsReturns read-only guidance for what to do next.No findings

    Why it is in this group

    • The name "list_next_steps" says "list": a word for reading.
    Description, as the server wrote it
    Returns read-only guidance for what to do next. Without campaignId, it returns the first-search question, active search progress, or campaign names and statuses to choose from. A single accessible campaign is resolved automatically. With campaignId, it returns a ranked to-do list: people who replied, people who accepted an invite but were never messaged, drafts and LinkedIn follow-ups waiting for approval, people with no reply for 7+ days while follow-ups are off, leads found but never contacted, and setup that is missing. It answers 'what should I do about this campaign today', where get_campaign_stats answers 'how is it doing'. For replies specifically, list_replies is the better source: its `awaitingOurReply` compares timestamps, while the `replied` bucket here covers only leads whose sentiment was scored and does not know whether we have since answered. `leadIds` passed to get_campaign_leads resolves the buckets into named people. Limits of the data: Onsa does not record whether we have already replied, or whether someone was contacted outside Onsa; 'replied' includes only replies whose sentiment was scored, so it is a floor; a withdrawn or unreachable invite leaves no trace, so some 'never contacted' leads may already have been tried. Sentiment is judged once per lead, not per message. Counts are complete; leadIds are capped at 200.

    Annotations the server declares

    • title: What to do next
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_pending_outreachLists outreach messages the agent has drafted that are waiting for a human to approve - the 'a message for X is ready' queue.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_pending_outreach" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "message": it reads messages.
    • untrusted-inputThe description's first sentence says "list" and "message": it reads messages, which other people can write.
    Description, as the server wrote it
    Lists outreach messages the agent has drafted that are waiting for a human to approve - the 'a message for X is ready' queue. Each entry carries the draft text, the lead it is for, and why that lead scored as it did. Without campaignId, the list covers every campaign in this workspace that the user takes part in. Each entry has `kind`: first_message, or follow_up (a LinkedIn follow-up for someone who did not reply, carrying the message it follows up in `followsUp`). This tool sends nothing: rewrite_outreach replaces a draft's text, and send_outreach queues one for delivery.

    Annotations the server declares

    • title: List pending outreach
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_repliesReturns the text of what prospects replied, for every lead in the campaign that answered, paired with the outbound message it answers.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_replies" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "message": it reads messages.
    • untrusted-inputThe name "list_replies" says "list" and "reply": it reads comments, which other people can write.
    • untrusted-inputThe description's first sentence says "return" and "message": it reads messages, which other people can write.
    Description, as the server wrote it
    Returns the text of what prospects replied, for every lead in the campaign that answered, paired with the outbound message it answers. get_campaign_stats counts replies and labels them; this returns the words. `sentiment` is Onsa's own label, written once per lead on their first reply - later replies never change it, and a reply Onsa has not scored yet comes back as `sentiment: null`, which means unscored rather than neutral. Those unscored replies are absent from get_campaign_stats entirely, so this tool can return more replies than the funnel counts. Three derived fields come with the reply. `awaitingOurReply`: the prospect spoke last and no sent message followed. It is structural only - a flat 'no thanks' satisfies it too - and Onsa sees only what Onsa sent, so a reply made by hand inside LinkedIn is invisible to it; what the data supports is 'no reply recorded here'. `daysSinceLastReply` is elapsed whole days rather than time-unanswered: it is populated even where we did answer, so it describes time-unanswered only when `awaitingOurReply` is also true. `looksLikeBroadcast`: the prospect's most recent message reached another profile in this campaign word for word, which indicates a mass DM rather than an answer. All three are floors rather than verdicts - a blast only one lead received is indistinguishable from a real reply, and two people who send the same long template are both flagged. Top-level `awaitingOurReplyCount` spans the whole campaign rather than this page, and leaves out broadcasts and replies scored negative; it can exceed `returned` when `limit` is small.

    Annotations the server declares

    • title: Read prospect replies
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • preview_shortlistPrepares a preview of 1–50 selected prospects from a campaign the user takes part in, and publishes nothing.No findings

    Why it is in this group

    • The name "preview_shortlist" says "preview": a word for reading.
    Description, as the server wrote it
    Prepares a preview of 1–50 selected prospects from a campaign the user takes part in, and publishes nothing. The snapshot holds a list title, which is the campaign's name, the note, and for each person only a name, role, company, LinkedIn profile and, when includeRationale is true, a fit rating and why-matched explanation. The preview returns the complete snapshot a link would expose, with the previewKey that create_shortlist requires; a created link stays viewable for 30 days by anyone who has it, and can be revoked with revoke_shortlist.

    Annotations the server declares

    • title: Preview a shareable list
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

48 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 2 outbound tools breaks every one.

Reads private data3
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write8
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out2
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__onsa__continue_campaign",
      "mcp__onsa__send_outreach"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:27 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
6
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://api.onsa.ai/.well-known/oauth-protected-resource/api/mcp
Authorization servers
https://api.onsa.ai
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
no
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 6 tools that change things (6 destructive). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://api.onsa.ai/api/mcp: tools/list returned 16 tools to an anonymous client, including create_shortlist (destructive), revoke_shortlist (destructive), find_leads (destructive), continue_campaign (destructive), rewrite_outreach (destructive) and 1 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn3 info
  • The server marks "find_leads" destructive (destructiveHint: true), and its name says "find", a word for reading.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "find_leads" says "find": a word for reading.

    What this check looks for
  • The input schema at properties.taskContext.description mentions "credentials".

    Matched text

    inputSchema.properties.taskContext.description: "...er profile, unrelated personal details, credentials, or guesses. This background does not a..."

    What this check looks for
  • The description is 3,755 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...size; each entry is a suggestion the user can accept, not something already done."

    What this check looks for
  • The input schema at properties.taskContext.description mentions "credentials".

    Matched text

    inputSchema.properties.taskContext.description: "...er profile, unrelated personal details, credentials, or guesses. This background does not a..."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http onsa https://api.onsa.ai/api/mcp
claude mcp add --transport http onsa https://api.onsa.ai/api/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__onsa__preview_shortlist",
      "mcp__onsa__fetch_leads",
      "mcp__onsa__list_campaigns",
      "mcp__onsa__get_campaign",
      "mcp__onsa__get_campaign_leads",
      "mcp__onsa__list_pending_outreach",
      "mcp__onsa__get_lead_memo",
      "mcp__onsa__list_replies",
      "mcp__onsa__list_next_steps",
      "mcp__onsa__get_campaign_stats"
    ],
    "ask": [],
    "deny": [
      "mcp__onsa__create_shortlist",
      "mcp__onsa__revoke_shortlist",
      "mcp__onsa__find_leads",
      "mcp__onsa__continue_campaign",
      "mcp__onsa__rewrite_outreach",
      "mcp__onsa__send_outreach"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as onsa. Review it before you use it.

Capability tags behind 8 rules
  • mcp__onsa__create_shortlist untrusted-input
  • mcp__onsa__revoke_shortlist untrusted-input
  • mcp__onsa__find_leads untrusted-input
  • mcp__onsa__continue_campaign untrusted-input, outbound
  • mcp__onsa__list_pending_outreach private-data, untrusted-input
  • mcp__onsa__send_outreach untrusted-input, outbound
  • mcp__onsa__list_replies private-data, untrusted-input
  • mcp__onsa__get_campaign_stats private-data, untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/ai.onsa/onsa/verify?hash=b009b3a661be23ee3c0950f35048484142051b6ce6738e7ed7b744342875211d" | jq -e '.changed == false'

Pinned hash b009b3a661be, from Nov 11, 2026, 20:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/ai.onsa/onsa/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for ai.onsa/onsa
[![Anlyon Radar](https://radar.anlyon.com/badge/ai.onsa/onsa.svg)](https://radar.anlyon.com/servers/ai.onsa/onsa)

Source: MCP Registry. Synced Oct 21, 2026, 17:42 UTC.

Onsa MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Onsa MCP server safe?

No scan can promise that. What its definitions show: 16 tools, 6 of which can change, send or delete something, and 4 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Onsa MCP server delete data?

Yes. 6 tools can delete, cancel or revoke something: create_shortlist, revoke_shortlist, find_leads, continue_campaign, rewrite_outreach and send_outreach.

Can the Onsa MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Onsa MCP server have?

16 tools: 6 destructive, 10 read. create_shortlist, revoke_shortlist, find_leads, continue_campaign, rewrite_outreach, send_outreach, preview_shortlist, fetch_leads and 8 more.

Does the Onsa MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Onsa MCP server send data to outside services?

2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): continue_campaign and send_outreach.

Has the Onsa MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 11, 2026.